Abstract
The main goal of this research was to determine the influence of the inner layer thickness and tightness, which depends on the length of the float determined by the float step and number of floats in the pattern repeat, on the thermal insulation and air permeability of the multilayered weft-knitted fabric. For this reason, weft-knitted spacer fabrics were produced by using woolen yarns for the outer layers and polyester filament yarn for the inner layer, with the number of floats in the inner layer varying between two and 20. Results from this research showed that the consequent increase in thickness and tightness of the spacer fabric’s inner layer has unequal effect on the thermal insulation and air permeability. Therefore, similar thermal insulation can be achieved by having significantly higher air permeability. This is especially evident over a long period of time.
Thermal insulation, particularly the dynamics of temperature changes through the textile, is an especially important characteristic of protective garments used in both very hot and very cold conditions. Required thermal insulation can be achieved by using a multilayered packet of textile fabrics or knitted multilayered or spacer fabrics. In the latter case, the duration of production and the number of production operations are reduced. Knitted spacer fabrics are characterized by a complex structure consisting of two outer layers joined together by an additional internal layer, which gives the fabric a spatial structure (3D). A 3D textile construction can be successfully produced with conventional knitting machines that are used for knitting 2D knits. As a result, this production method of weft-knitted spacer structures has recently become particularly attractive due to the simple equipment and maintenance, wide application of patterns, and raw materials. It is well known that the knitting pattern and raw material of yarns are the main influence on the physical and mechanical properties of knitted fabrics. A knitted spacer structure, due to its thickness, is particularly suitable for thermal insulation. Thermal insulation strongly correlates with the geometrical and structural characteristics of a knitted fabric, as well as the raw material and structural parameters of yarns used; however, it was found that the influence of the pattern and structural-geometrical characteristics of the knit on heating and cooling dynamics is higher than that of the raw material. 1 Thermal properties are important for wearing comfort and for protection against the heat. Protective garments must protect humans from environmental hazards, allow functions of the body, and ensure good comfort properties, while protection, together with comfort, is important for the working efficiency and number of accidents occurring during work. 2
Thermal properties of textile materials and clothing are considered to be one of the most important features due to the high demand for comfort, and there are various factors affecting comfort, such as air permeability, thermal conductivity, thermal absorptivity, and water vapor permeability.3,4 Thermo-physiological comfort helps the human body to retain heat balance while at rest or during various levels of physical activity. 5 Thermal comfort capacity plays an important role in the development of textile garments, and is mainly conducted in aspects of air and vapor permeability and heat transfer properties. 6
The phenomenon of thermal comfort is related to many factors which characterize a human body, the surrounding environment, and the clothing that humans wear.6–8 An exchange of thermal energy between a human and the surrounding environment proceeds continually and the process preserves this continuous dynamic balance.9,10 Thermal insulation is provided by the air trapped between fibers and yarns, as well as between several layers of multilayered fabric. 11 Thus, in order to increase thermal insulation, the amount of air gaps between layers has to be increased too. 12 The dynamics of heat transfer from the human body to the surrounding environment is an important functional and comfort property. It correlates on some level with air permeability, which is also recognized as one of the most important properties of comfort.12–16 Airflow through textiles is mainly affected by pore characteristics, the dimension and distribution of which are a function of fabric geometry. The air permeability of textiles has been widely analyzed by many researchers and it is well known that it depends on such factors as structural parameters of the fabric, porosity, thickness, density, finishing, linear density of yarns, and composition of the yarn.17–21
Previous research has shown that the thermal properties of textile fabrics depend on the fabric’s thickness and porosity, therefore, fabrics knitted in combined patterns have higher thermal resistance than plain plaited fabrics because of the higher thickness of the combined structures. 22 As the thickness of fabrics knitted from the same raw material increases, thermal resistance also increases, which means that thicker knitted fabrics are warmer. 22 Therefore the highest thermal insulation can be achieved by using knitted double-layered or spacer fabrics.
In the previous research, it was found that weft-knitted spacer structures, which consist of two outer layers connected by floats of yarn laid between two needle-beds in the inner layer, give good heat insulation and dynamics over time. 3 Furthermore, thermal insulation of such structures strongly correlates with the raw material of the yarns used for knitting particular layers. The best thermal insulation was obtained by using woolen yarns for the outer layers and polyester filament yarn for the inner layer, and the knitted structure for various combinations of raw materials of all investigated fabrics was the same. 3
The main goal of this research was to determine the influence of the inner layer tightness and thickness, which depends on the length of the float determined by the float step and the number of floats in the pattern repeat, on the thermal insulation and air permeability of the weft-knitted spacer fabric.
Materials and methods
Six variants of knitted spacer structures were developed for this research. Experimental samples were knitted on the E12 gage, flat double needle-bed knitting machine SES 122-S (Shima Seiki, Japan). After knitting, all samples were stabilized by using steam thermo-stabilization equipment Cosmotex (Spain). For all variants, the same raw composition was used. Folded woolen yarns with 33.3 tex ×2 linear density, 210 m−1 twist level and z twist direction (Filivivi Srl, Italy) were used for knitting both outer layers, while polyester filament yarn (PES) with 42 tex linear density/167 filaments (Lakshmi Ganapathy Textiles, India) was used for knitting the inner layer. Such raw composition was chosen due to the well-known good thermal characteristics of woolen yarns, while the filament polyester yarn keeps the dimensional shape of the multilayered structure and ensures more air gaps between the yarns. 3
The developed knitted spacer structures differ in pattern repeats of the inner layer, namely in the step of inner layer floats which determines their length, and the number of the floats in the pattern repeat. The highest float step was nine needles, and each second structure has floats with a two-needle lower step. Thus, the lowest step is zero, where the float connects two tucks formed on two adjacent needles of the front and back needle-bars. The real view of the fabrics is presented in Figure 1. Knitting structures of all six developed multilayered fabrics are presented in Figure 2.

Images of the knitted spacer fabrics: (a) S9 (float step is nine needles); (b) S7 (float step is seven needles); (c) S5 (float step is five needles); (d) S3 (float step is three needles); (e) S1 (float step is one needle); (f) S0 (float step is zero needles)

Knitting structure of the knitted spacer fabrics: (a) S9; (b) S7; (c) S5; (d) S3; (e) S1; (f) S0
The main characteristics of the tested knitted fabrics are presented in Table 1.
Characteristics of weft-knitted spacer fabrics
Structure parameters of knitted samples were analyzed according to the British Standard BS 5441:1998. Values of the wale and course density on the technical face and technical back sides were similar. Thickness was measured according to the standard EN ISO 5084:1996, with a 0.001 mm measurement error. The coefficient of variation of the thickness measurements did not exceed 4.2%.
All experiments were carried out in a standard atmosphere for testing according to the standard ISO 139:2002.
Air permeability of the knitted samples was measured using an L14DR device (Karl Schroder KG, Germany) according to Standard LST EN ISO 9237:2007. The air flow was measured in the circle-shaped area of 5 cm2 at 200 Pa pressure. There were 20 tests performed for each experimental point. The absolute error of the measurements was calculated with a confidence level of 0.95. The air permeability values were calculated according to the formula:
The dependence of heat exchange on the structure of the inner layer of the knitted spacer fabrics was investigated using the IG/ISOC (Giuliani Technologies, Italy) attachment designed for investigation of the heat insulation. The sample of knitted fabric was laid down on the heated plate, technical back side down. The temperature of the fabric surface was measured using a digital thermometer, HD9214, with a platinum sensor, PT100 (DELTA OHM SRL, Italy), superimposed on the outward side of the fabric. The measurement error of the digital thermometer with a platinum thermo-sensor is equal ± 0.071 - 0.076 °C. The plate was heated up to 40°C and this temperature was maintained during the experiment. The environmental temperature was also kept constant at 25°C. The changes of temperature were observed for a 3600 second (1 hour) period and recorded every 10 seconds. Five tests for each experimental point were performed.
Results and discussions
It is well known that the thickness of a fabric may influence other physical properties, such as air permeability and thermal insulation. However, correlation between the fabric thickness and air permeability can only be objectively analyzed when the raw composition and yarn structure for all compared fabrics are the same. 3 Analysis of the newly developed weft-knitted spacer fabrics showed that fabric thickness continuously decreased by decreasing the number and step of floats in the inner layer. The lowest difference of thickness is obtained between the fabrics S0 and S1, only 5% (see Table 1), while the difference between the fabrics S1 and S3 thickness is 46%, and is the highest difference between all investigated fabrics. Images of spacer fabrics presented in Figure 1 demonstrate that the thickness of the inner layer of the specimens S0 and S1, produced with four and eight floats, respectively, is not significant. Furthermore, the total thickness of the fabrics S0 and S1 is similar at 3.5 mm and 3.7 mm, respectively. Starting from the fabric S3 with sixteen floats in the pattern repeat (see in Figure 2 and Table 1), the fabric thickness significantly increases. When the float step increases from three to five (fabrics S3 and S5, respectively), the increase of the thickness is 29%. An increase in float step from five to seven (fabrics S5 and S7, respectively) and from seven to nine (samples S7 and S9, respectively) gives approximately a 15% higher fabric thickness. Thus, the fabric thickness was determined by the number of floats in the pattern repeat of the inner layer; the higher the number of these floats in the pattern repeat, the greater the thickness of the knitted fabric. However, the consistent increase in the float step and number of floats in the pattern repeat gives an uneven increase of the fabric thickness. When the tightness of the inner layer is relatively low, as it is for fabrics S0 and S1, or the tightness of the inner layer becomes high, as it is for fabrics S7 and S9, the influence of change of the number of floats is not as significant as for fabrics S3 to S7.
The main goal of this research was to establish the influence of the inner layer structure of the weft-knitted spacer fabric on the air permeability and thermal properties such as thermal insulation and heat exchange dynamics through the fabric.
From the data presented in Figure 3, it is evident that the heat exchange dynamics through the knitted spacer fabric strongly depends on the structure of the inner layer, i.e. on the number of floats in the inner layer and float step used in the pattern repeat. The obtained results show that dependence of the thermal insulation and heat exchange dynamics through the spacer fabrics on the inner layer pattern repeat decreases over time, and there can be distinguished some similar groups of the fabrics. Structures S0 and S1, with the shortest floats (lowest float step) and the lowest number of floats in the pattern repeat, depend on the first group. These fabrics showed similar heat exchange dynamics (especially in the first period of observation, 0.3°C difference appears only after 30 minutes) as well as the lowest thermal insulation in comparison with all investigated samples. The thermal insulation continuously increases with the increase of the float step in the pattern repeat, which gives longer floats and numbers of floats in the pattern repeat. The difference between thermal insulation of the S1, S3, S5, and S7 structures in the whole period of observation is significant; approximately 1.0-1.5°C. Fabrics S7 and S9 also presented similar results, despite the fact that the difference in the pattern repeat of the inner layer is the same as for the other mentioned fabrics (S1 to S7). The difference in thermal insulation between fabrics S7 and S9 is only approximately 0.3°C, i.e. in the margins of error.

Dependence of heat exchange dynamics on the structure of the inner layer of the knitted spacer fabric
More detailed analysis can be done from the data presented in Table 2. In this table, results of the temperature reached on the upper layer of the weft-knitted spacer fabrics after 1 min, 15 min, 30 min, and 1 hour are presented. It is important to know how much these materials can protect in long-term conditions. At the start (0 s), the temperature on the upper surface of all investigated spacer fabrics was 25°C, equal to the ambient temperature. The first small difference appeared after the first minute of observation. The temperature on the upper layer of fabric S0 reached 26.8°C and fabric S1 reached 26.5°C, while the temperature on the upper layer of fabrics S3 to S9 was the same, 26°C. A more significant difference in temperature appeared between three fabric groups (S0-S1, S3-S5, and S7-S9 after 15 min), while the difference within each group was only up to 0.4°C. After 30 min, the difference between the temperature on the upper surface of the S1 and S3, S3 and S5, S5 and S7 multilayered fabrics was more than 1°C, while the difference between structures S0 and S1, and S7 and S9 was only 0.3°C, i.e. in the margins of error. As expected, the best thermal insulation was obtained for the S9 structure with the highest float step and number of floats in the inner layer. After 1 hour, the temperature on the upper layer of fabric S9 was 33.7°C, more than 6°C degrees lower than the heating plate. However, a similar temperature (34°C) was also reached on the upper surface of fabric S7. The difference in the temperature between fabrics S7, S5, S3, and S1 was significant, approximately 1°C, while the difference in temperature between fabrics S0-S1 and S7-S9 remained only 0.3°C.
Temperature of the upper layer of the heated spacer fabrics depending on the structure of the inner layer
Thus, it is evident that thermal insulation during a long-term period strongly depends on the inner layer structure of the knitted spacer fabric. However, the significance of the temperature differences between fabrics with a consistently changed float step and number of floats in the pattern repeat is unequal. It means that, in some cases, similar thermal insulation can be achieved with lower consumption of yarns in the inner layer and lower thickness of the fabric. Dependence of the thermal insulation on the fabric’s thickness after a particular time of observation is presented in Figure 4. There is a strong dependence between the temperature and thickness (coefficient of determination R2 is more than 0.98). The obtained results demonstrate that the influence of the thickness increase on thermal insulation of thicker spacer structures is not as significant as for fabrics with lower thickness. A 15% increase in the thickness (between fabrics S7 and S9) does not give significant change in thermal insulation. In all presented periods this difference is up to 0.3°C. It demonstrates that thickness and tightness of the inner layer reached a level which gives better thermal insulation characteristics for such spacer structures, and further increases in fabric thickness may only have a significant effect on the thermal insulation after a much longer period of at least a few hours.

Dependence of temperature on the thickness of the knitted spacer fabric
In order to investigate any influence of the inner layer structure and thickness of the newly designed weft-knitted spacer fabrics on air permeability of these fabrics, air permeability tests were performed. Air permeability values of the knitted spacer fabric, depending on the float step and number of floats in the pattern repeat, are presented in Figure 5.

Air permeability of knitted spacer fabrics with different structures of the inner layer
As can be seen from the data presented in Figure 5, air permeability decreases with increasing float steps and the number of floats in the pattern repeat of the inner layer. Fabric S0, with the four shortest floats in the pattern repeat of the inner layer, has the highest value of air permeability (801.60 dm3/m2s), while fabric S9, with 40 floats of nine-needle steps in the pattern repeat of the inner layer, has the lowest air permeability (517.70 dm3/m2s). Dependence of air permeability on the spacer fabric thickness is graphically presented in Figure 6. It demonstrates that air permeability strongly depends on the fabric thickness. However, such dependence can be analyzed only when compared fabrics are knitted using yarns of the same raw material and the same characteristics (linear density, twist level, surface hairiness, the same spinning system, etc.) 3 From the obtained results it is evident that continuously increased float steps (from zero to seven) and the number of floats in the pattern repeat (from four to 32) give approximately a 9-11% decrease of the air permeability. However, when the tightness of the inner layer of the spacer fabric reaches a particular level, the influence of the fabric thickness on the air permeability significantly reduces. Fabric S9 has 15% higher thickness than S7, however, air permeability of fabric S9 is only 2.8% lower than that of fabric S7. It can be explained by the fact that the tightness of the inner layer of the higher pattern repeat is so high that further increases do not have as much of an influence on changes in the air permeability.

Dependence of air permeability on thickness of knitted spacer fabric
As there is a strong dependence of the thermal insulation on fabric thickness as well as of the air permeability on the thickness, there was an investigation into whether a correlation exists between air permeability and thermal insulation (in various periods of observation after 15 min, 30 min and 1 hour) of the weft-knitted spacer fabrics. This correlation is presented in Figure 7.

Correlation between thermal insulation and air permeability of knitted spacer fabric
It was found that a strong linear correlation exists between the thermal insulation and air permeability (R2 is more than 0.93), and is due to the same yarns and the same loop density of the outer layers being used for knitting the investigated knitted spacer fabrics. 3 Thermal insulation is inversely proportional to the air permeability, so higher thermal insulation corresponds to a lower air permeability of the fabric. However, different zones of correlation between the thermal insulation and air permeability can be identified in Figure 7. In the first zone of fabrics with high tightness and thickness of the inner layer, a 0.3°C difference between the thermal insulation of fabrics S9 and S7 after 1 h of observation corresponds to a 25 dm3/(m2·s) difference in the air permeability. In this case, a change in both the air permeability and thermal insulation are insignificant. It means that S7 fabric, knitted with a lower yarn consumption for the inner layer, has similar thermal insulation and comfort (air permeability) properties as fabric S9. In the third zone of fabrics with relatively low thickness and tightness of the inner layer, the same 0.3°C difference of thermal insulation between fabrics S0 and S1 corresponds to an 80 dm3/(m2·s) difference of their air permeability. In this case, similar thermal insulation over a long-term period can be achieved by having significantly higher air permeability. In the middle zone, comparing fabrics S1 to S7, it can be seen that significant changes in thermal insulation (1°C and more) over a long-term period also corresponds to significant changes of the air permeability (53 – 64 dm3/(m2·s)). In this case, the higher number of floats in the pattern repeat of the inner layer and higher length of the floats determines significantly higher thermal insulation and lower air permeability.
It can be concluded that a consequent increase in thickness and tightness of the spacer fabric’s inner layer, depending on the float steps and number of floats in the pattern repeat of the inner layer, has unequal influence on the thermal insulation and air permeability and should be analyzed at the level of design in order to obtain optimal thermal insulation with the lowest consumption of yarns and the highest air permeability.
Conclusion
It can be concluded that all developed weft-knitted spacer fabrics can be used for thermal insulation as the temperature on the upper surface of all investigated structures after 1 h did not reach 40°C, i.e. the temperature of the heating plate on which the fabrics were laid for heating.
The increase of the float step from zero to nine needles and an increase in the number of floats in the inner layer of the knitted spacer fabrics from four to 40 determined the increase of the fabric thickness from 3.5 mm to 10 mm. It was found that thermal insulation and heat exchange dynamics through the spacer fabric strongly depends on the inner layer thickness of the fabric, however, the significance of the temperature exchange differences between the fabrics with consistently changing float steps and number of floats in the pattern repeat is unequal. The influence of the thickness increase on the thermal insulation of thicker multilayered structures is less significant than for fabrics with lower thickness. A 15% increase of the thickness between fabrics with 40 nine-needle step floats and 32 seven-needle step floats in the pattern repeat gives only a 0.3°C change in thermal insulation. While comparing fabrics with one-to-seven-needle steps and eight to 32 floats this difference is 1°C and more.
Also, a strong linear correlation was found between thermal insulation and air permeability; however, different zones of correlation can be identified. For the thicker and tighter spacer fabrics (with 40 nine-needle step floats and 32 seven-needle step floats in the pattern repeat), only a 25 dm3/(m2·s) difference of the air permeability and 0.3°C difference after 1 h was found. Thus, a similar thermal insulation can be achieved with lower consumption of yarns in the inner layer and lower thickness of the fabric. The same 0.3°C difference in thermal insulation and 80 dm3/(m2·s) difference of the air permeability corresponds to the fabrics with four zero-needle steps and eight one-needle step floats in the pattern repeat. In this case, similar thermal insulation over a long-term period can be achieved by having significantly higher air permeability.
Footnotes
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
